While vaccine strategies for the generation of tumor specific immunity in humans continue to have great promise, to date they have been less than overwhelming in their antitumor efficacy. Studies supported by this award pioneered the use of intralesional therapy using cytokine encoding vaccinia recombinants for the immunologically based gene therapy of cancer. Following preclinical validation, we translated these finding to the first clinical trials of this strategy which resulted in the encouraging result of 4/7 melanoma patients treated having non-injected lesions regress with 2 long term remissions (1) We have identified as a principal limitation of this and other approaches, the downregulation or redirection of tumor immune responses by tumor production of one or several suppressive molecules. In some cases factors such as VEGF and TGFbeta have direct and indirect tumor growth enhancing activities as well as immune suppressive activities while others such as IL10 paradoxically have immune regulatory activity without known pro-tumor functions. The studies outlined in this competitive renewal application will explore the antitumor and immunologic effects of modulating these suppressive mechanisms on the efficacy of our intralesional as well as antigen-encoding recombinant vaccinia vaccine strategies in tumor-bearing mice. For the first time, we will compare the antitumor and immune responses following intralesional gene transfer using cytokine constructs with the activity of antigen-encoding poxvirus vectors in two murine models which express overlapping combinations of suppressive molecules. We will then examine the effects of systemically and locally blocking the activities of the suppressive factors at the tumor site in the case of our intralesional strategy and the vaccine site in the case of the antigen-encoding vaccine. Systemic modulation will utilize antibodies and small molecules while our local studies both intralesional and at the vaccine site will utilize a series of molecularly engineered fusion molecules designed in our laboratory and delivered using our poxvirus vectors. We have specifically chosen this dual-pronged approach based on our and others' demonstrated clinical utilization of vial recombinants and the resultant feasibility of translating the results rapidly to clinical trials. By comparing the two vaccine approaches and the effects of modulating the "immunization site" expression of suppressive molecules, we should both significantly advance our localized treatment strategy as well as perhaps enhance antitumor vaccine approaches using defined antigens. Finally, we will attempt to consolidate our intralesional immunization with antigen-encoding vaccines that will give us a means of boosting responses in the absence of accessible tumor.